10 years of creating 1,000 pieces of content… Spending $15k on courses and equipment…
And those 1,000 pieces of content got less then 100 views…
This taught me 1 very… very… painful lesson…
I don’t know what good content is.
And if your content isn’t gaining traction, neither do you.
For years, I thought “good content” meant well-written, passionate, authentic posts that came from the heart.
I was dead wrong.
Good content is content that moves people to action.
Clapping. Saving. Commenting. Sharing. These are signs that your content moved your audience. This is why every algorithm incentivizes these in some way.
If your audience isn’t doing any of these things, your content isn’t good — no matter how much you love it.
This realization nearly made me give up creating content.
I don’t know how to make content that moves people… So… am I f*cked?…
Thankfully, I can speak from personal experience here: No, you’re not.
Your problem isn’t your work ethic. It’s not your platform. It’s not even your “voice.”
It’s that you’re treating content like art when it’s actually data.
The Lie We’re All Told
So, how did we get here?
How did we end up believing that our passion and authenticity were enough?
Because we were told a well-intentioned lie.
The gurus, the experts, the thought leaders… they all say the same advice:
- “Find your passion.”
- “Just be authentic.”
- “Post consistently.”
For 1,000 articles, I followed their advice to the letter. I was passionate. I was authentic. I was consistent. And it led me straight to a digital graveyard.
That advice is a trap. It encourages activity over strategy.
For years, I was trying to be a creative genius. I told myself I had a unique “style” and that learning frameworks or studying data would ruin it.
The truth is… I was hiding.
I was like a musician trying to improvise without ever learning the rudiments. My 1,000 failed pieces of content were the sound of me banging on the drum set, hoping to accidentally write a perfect song.
I hadn’t earned the right to have a style because I hadn’t mastered the fundamentals of what actually works.
I had to learn my “rudiments” first. And those rudiments are built on a single, powerful shift in mindset.
From Content Library to Content Laboratory
That single shift was this:
I stopped trying to build a library and started building a laboratory.
In a library, every piece needs to be a polished, timeless classic. The pressure is immense. Every article feels like you’re writing a book.
In a laboratory, 99% of your experiments are supposed to fail. The failure is the data. The failure is the point.
This is how you solve the problem of not knowing what “good” content is. You stop guessing. You let the audience’s actions — the data — tell you what’s good.
Part of my lab work was dissecting my own failed content.
Here’s an example of what I learned:
I had heard “people connect with stories… start with a story to bring people in.”
So nearly 1/2 of my 1,000 pieces of content started with a cool story or a neat idea. But they were pointless. They had no steaks.
The breakthrough came when I learned to work backward.
I had to start building the content around the Antidote: the one single, valuable takeaway for the reader. Only then could I build a story around it.
This simple shift meant every “experiment” was now designed to deliver value, even if it didn’t go viral. It was a tool for designing better tests.
And this process is the foundation for how a Content Laboratory actually works.
How the Content Laboratory Actually Works
Your goal for the next 50 articles is not to get followers. It is to get data.
Your goal is to invalidate as many bad ideas as possible, as quickly as possible. Every dead-end you find is a victory.
Here’s the system:
Step 1: The Hypothesis (For a Specific Person)
Stop writing about “topics.” Start testing a thesis.
A thesis is a specific, falsifiable statement about a specific person’s pain.
In my old model, I’d write a generic post about “writing tips.” It would fail every time.
In the lab, I would create a specific hypothesis for a specific person.
Here’s an example:
My person: Jason, the Expert Writer on Medium. He pours hours into well-researched articles, but they die with less than 100 views. He’s frustrated and feels invisible.
My thesis could be: Writers like Jason are frustrated by low views (Pain) because their headlines are framed around the topic, not the reader’s curiosity (Problem). They will value a simple framework for writing headlines that get clicks (Solution).
This experiment wouldn’t be another generic “how-to” article. It’s a Minimum Viable Article: “A 3-Step Framework to Write Headlines That Medium’s Algorithm Can’t Ignore.”
The thesis would be proven correct if the data showed a spike in views. Views = someone clicked on the article. The headline is one of the core elements that causes someone to click, so higher views means this article idea structure a nerve.
- Before (Topic): “I’m going to write about how to be a better writer.”
- After (Thesis): “I believe expert writers on Medium (Person) are frustrated by low views (Pain) and will respond to a specific, tactical framework for writing better headlines (Solution). This article is my test.”
Step 2: The Minimum Viable Article (MVA)
My 1,000 failed pieces taught me that spending 10 hours on an untested idea is malpractice.
An MVA is a ruthlessly efficient test. It has a clear idea, a decent hook, and a clean structure. Its only job is to prove your hypothesis right or wrong.
So how can you run experiments quickly without burning out?
Stop being a lone genius.
Some people are going to unfollow me for this, but you need to treat AI as my “eager intern.” I have a custom-trained agent on my writing voice, and I give the agent my core idea, the raw data from a voice recording (more on this idea here), and the target person I’m writing for. The AI does the grunt work — it asks me questions to clarify points, outlines the structure, writes the first draft, and does the formatting.
This frees me up to do the work of the lead scientist: refining the thesis, injecting my unique voice, and analyzing the results.
This system is how I went from spending 10 hours on a failed article to 90 minutes on a valuable data point.
Step 3: Analyze the Data (The Algorithm as Your Lab Assistant)
For 1,000 pieces of content, I treated the algorithm like a casino slot machine, hoping to hit the jackpot. That was wrong.
The algorithm is your lab assistant. And you have to train it.
My 1,000 failed pieces of content were a terrible curriculum, teaching it that I was a random content creator for random people. When I started running clean experiments for Jason, the algorithm started learning. It said, “Ah, I see. You make content for this type of person.”
Every successful experiment isn’t just a win for you; it’s a lesson that makes your assistant smarter.
Here are the metrics that actually matter:
- Views = A Successful Package. As we said in Step 1, a spike in views means your headline and topic resonated. The packaging worked. You got them in the door.
- Read Ratio > 60% = A Strong Hook. They clicked, but did they stay? A high read ratio means your introduction was strong and the core idea was compelling.
- Saves = Product Demand.
This is the single most important metric. It means your idea is so valuable people are afraid to lose it. This is what you build a business on.
- Comments = A Strong Stance. You’ve found a polarizing idea that builds community and gets people talking. Double down on this perspective (assuming you actually believe what you say, and are not rage-baiting people).
- Silence = A Failed Experiment. A beautiful, valuable, data-rich failure. The hypothesis was wrong. Kill it without emotion and move on to the next test.
Want to go deeper into exploring what each metric means? I have a 21 day email series called How to Train the Algorithm, that goes DEEP into the meaning behind the metrics. Click here to join it (free).
Step 4: Iterate or Kill
Based on the data from your lab assistant, you have two choices. (This is where you really stop being an artist and start being a strategist).
1. Iterate
If an idea shows a spark of life, you iterate.
A “spark of life” isn’t a viral hit. It’s a signal that something worked. Maybe you had a high read ratio but low views (great idea, bad packaging). Or maybe you had high views but no saves (great packaging, weak idea).
When this happens, you don’t throw the idea away. You change one variable and run the experiment again.
- Re-package it with a stronger, more specific headline.
- Re-frame the intro to create a better hook.
- Add more data or a better story to make the core idea more valuable.
The goal is to isolate what worked, amplify it, and test again.
2. Kill
If an idea gets zero signal across multiple tests, you kill it.
Without emotion. Without mercy.
Silence is data. It’s the market telling you, “We don’t care about this.” Your job is to listen.
Killing a failed hypothesis isn’t a failure. It’s a victory. You’ve successfully invalidated an idea, which saves you hundreds of hours you might have wasted on it.
Deleting it is a good thing — it cleans up your lab and stops confusing your lab assistant, the algorithm.
The Content Laboratory at work
I still publish things that “fail.” But they’re not failures. They are intentional, low-cost experiments that tell me where to go next.
Stop treating content as a final product and start treating it as a tool.
This is how you spend less time on better content.
(Wanna get meta? This article you’re reading is a reframed version of an idea I wrote about on my email list, that got lots of opens and replies. You can check out the original idea here).
You stop guessing what people want, and you build a system that makes them show you. You don’t spend six months building a course nobody wants. You spend six weeks testing ideas with MVAs, and by the end, the market has told you exactly what product to build based on what they saved, commented on, shared, etc.
You’ve used data to turn your wisdom into a validated business idea.
Serve first, profit follows.
You don’t need to write 1,000 articles to learn this lesson.
Your first 50 articles aren’t a library. They’re a laboratory.
Start experimenting.
P.S. Ready to stop guessing and start experimenting?
The entire system — from generating a thesis to analyzing the data — is what we practice inside the Writerpreneur community. If you want the frameworks, ideas, and workflows that turn your content into a data-driven business, this is the place to start.
Because your knowledge has value — you just need the right system to prove it.
